Effects of vaginal vs oral progesterone supplementation before embryo transfer on live birth rates and levels: a randomized trial.

This trial was supported by Basque Government and included in HAZITEK program, framed in the new Euskadi 2030 Science and Technology Plan (PCTI 2030). These aids are co-financed by the European Regional Development Fund (FEDER). The study funders had no role in the study design, implementation, analysis, manuscript preparation or decision to submit this article for publication.

The baseline features of patients were similar between the randomized groups. The flowchart of the patients entering the trial is shown in Fig. 1 . Both groups were comparable in age, years of infertility, ovarian reserve and number of previous ART cycles. There were no differences in clinical pregnancy (50.20 vs 47.37%), live birth (43.67 vs 40.89%) or miscarriage rates (13.33 vs 6.48%) depending on the route of progesterone administration before embryo transfer (Pre-ET) ( Table 1 ). However, the plasmatic levels of progesterone on the day of ET (PP1) were significantly higher when pre-ET progesterone administration was oral (group A) compared to the vaginal route (group B) (35.10 ± 5.61 vs 15.13 ± 2.73 ng/mL; t = 1.559; P = 0.002). Clinical, analytic and ultrasound parameters. Comparative data according to pre-ETprogesterone administration schedule. BMI, body mass index. Plasma progesterone levels on the day of embryo transfer (ET) (ng/mL). Plasma progesterone levels on the day of B-HCG determination (ET) (ng/mL). Includes one ectopic (tubal pregnancy). All 208 newborns were available for analysis, including sex, weight and type of delivery. There were no differences in the percentage of preterm deliveries and newborns’ mean weight between the study’s two arms ( Table 1 ). There was a higher percentage of cesarean deliveries among the patients who received vaginal progesterone before embryo transfer (group B) (45.55 vs 28.97%; chi-square = 6.12; P = 0.013). The proportion of females/males was higher among group A patients (oral pre-ET progesterone) ( Table 1 ). Progesterone levels on the day of embryo transfer (PP1) were significantly higher among patients with evolutive pregnancies than among those without evolutive pregnancies and no pregnancies ( Table 2 ). The P1 levels were similar between patients with non-ongoing pregnancies (including miscarriages and non-ongoing positive pregnancy tests) and non-pregnant patients ( Table 2 ). Comparative progesterone determinations and endometrial thickness data. EP, evolutive pregnancy. Plasmatic progesterone levels on the day of embryo transfer (ET) (ng/mL). Plasmatic progesterone levels on the day of B-HCG determination (ng/mL). Evolutive pregnancy included full-term and preterm deliveries (LBR). Included miscarriages and non evolutive positive pregnancy test. To go further, progesterone values on the ET day (PP1) were examined dichotomously, stratified by the 10th, 25th and 50th percentiles. It was noted that while the 10th percentile lacked discriminatory ability, significance was observed at the 25th and 50th percentiles ( Table 3 and Fig. 3 ). Ongoing (live birth) pregnancies according to plasmatic progesterone levels on the day of embryo transfer (PP1). Comparative results between values below or above are 10th, 25th and 50th percentiles. PPI-ET, plasmatic progesterone levels on day of embryo transfer (ET) (ng/mL); OPR, ongoing pregnancy rate. LBR according to progesterone levels stratified by percentiles on the day of embryo transfer (PP1). An area under the curve (ROC) was analyzed to determine the potential predictive value of progesterone levels on the day of embryo transfer in a full-term pregnancy. The PP1 levels did not predict an evolutive gestation (AUC 0.491 with a 95% confidence interval between 0.380 and 0.601) ( Table 5 ). Regarding secondary outcomes, progesterone levels on the day of B-HCG determination (PP2) were not associated with pregnancy success or miscarriage rates ( Tables 2 and 4 and Fig. 4 ). Ongoing (live birth) pregnancies according to plasmatic progesterone levels on day of pregnancy test (PP2). Comparative results between values below or above are 10th, 25th and 50th percentiles. PP2_B-HCG, plasmatic progesterone levels on the day of B-HCG determination (pregnancy test) (ng/mL); OPR, ongoing pregnancy rate. Live birth rate according to progesterone levels stratified by percentiles on the day of B-HCG determination (PP2). Endometrial thickness showed significant differences when comparing ongoing pregnancies to non-ongoing pregnancies. However, no significant differences were observed between ongoing pregnancies and non-pregnancies ( Table 2 ). By assessing ROC curves, the predictive value of endometrial thickness for determining progressive pregnancy was not statistically significant ( Table 5 ). Area under the curve (AUC) (ROC) (parameters related to live births). Null hypothesis: area = 0.5. A-CI, asymptomatic confidence interval; PP1-ET, plasma progesterone levels on the day of embryo transfer; PP2_B-HCG, plasma progesterone levels on the day of B-HCG determination.

This was a prospective, randomized, controlled study performed between September 2022 and June 2023 at the University-associated Assisted Reproduction Center to properly investigate the potential influence of different progesterone administration protocols before embryo transfer and plasmatic progesterone levels on implantation and pregnancy rates in frozen euploid embryo transfers. After ovarian stimulation and egg retrieval, a blastocyst trophoectoderm biopsy and cryopreservation of all embryos were performed. Since only single embryo transfers of frozen euploid blastocysts are performed at our center, no double embryo transfer of non-screened embryos was included. The clinical trial was approved by the Institutional Ethics Committee of Clinical Research of Euskadi (CEIC-E) and the Institutional Review Board of the University of the Basque Country (UPV/EHU (Clinical Trial Number EUDRACT: 2022-000382-41) on August 20, 2022, and the AEMPS (Agencia Española de Productos Sanitarios y Medicamentos; Spanish Agency for Medicines and medical devices). The study was conducted in compliance with Good Clinical Practice guidelines. CONSORT reporting guidelines have been used ( Schulz et al. 2010 ). This study was approved in accordance with the ethical principles that have their origin in the Declaration of Helsinki and Ethical Guidelines for Biomedical Research on Human Participants. All patients received and signed a written informed consent form. Patients with available euploid blastocysts derived from own oocytes after PGT-A were included in the trial. All women reported regular length of the menstrual cycle. Exclusion criteria were age ≥42 years, body mass index (BMI) above 32 or less than 18, endocrine or autoimmune disease (e.g., diabetes, thyroid disease or presence of anti-thyroid antibodies or PCOS), chromosomal abnormality, uterine malformations, history of intra-abdominal infectious pathology or surgery, diagnosis of endometriosis, adenomyosis, uterine fibroids, absolute or relative contraindication for a follicular puncture and the use of testicular sperm for ICSI. Couples involved in other clinical or embryological trials were also excluded. Embryo transfers from donated oocytes were excluded. A total of 500 with planned embryo transfers of vitrified-warmed euploid blastocysts were invited to participate in this randomized controlled trial (RCT) and were randomized 1:1 into two groups: group A, in which progesterone before embryo transfer (pre-ET) was administered orally, and group B, with patients who received pre-ET progesterone vaginally. Of these, 492 patients ultimately reached study completion ( Fig. 1 ). Randomization chart of the trial and trial design. A block randomization process was performed before initial treatment. Patients were allocated to the mentioned treatment groups using the computer-generated randomization codes, which were then placed in the sealed, opaque, sequentially numbered envelopes by a third party (nurse practitioner), who was not directly involved in the patient management or the randomization process. The envelopes were handed out to the participants upon completing the informed consent. Both the patients and the clinicians were aware of the allocated arm. The participants were followed through one completed ART cycle until all frozen embryos generated from the index cycle were transferred or until delivery in those who achieved pregnancy. All patients underwent ovarian stimulation, egg retrieval and ICSI procedures described elsewhere ( Barrenetxea et al. 2024 ). A GnRH antagonist protocol was scheduled for ovarian stimulations. A single injection of corifollitropin alpha (Elonva®, Organon NV, The Netherlands) on day 2 of the menstrual cycle was followed by the administration of recombinant follicle-stimulating hormone (rFSH) (Bemfola 225–300 IU/day, depending on ovarian reserve, Gedeon Richter Plc. Budapest Hungary). GnRH antagonist (Ganirelix 250 μg/day, Organon NV, The Netherlands) was started when a leading follicle of 14 mm was achieved. A GnRH agonist (Decapeptyl 0.2 mg, Ferring Pharmaceuticals, Switzerland) trigger was administered when at least one follicle was above 18 mm. Ultrasound-guided transvaginal oocyte retrieval was performed 36–38 h after the trigger injection using an 18-gauge double-lumen needle and a vacuum pump (Cook Medical Incorporated, USA) under pressure at 150 mmHg. Follicle flushing was not performed. The cumulus–oocyte complexes were removed from the collection fluid using a sterile glass pipette and washed in G-IVF Plus media (Vitrolife, Sweden). All embryos were cultured until the blastocyst stage. A trophectoderm biopsy of developed blastocysts was performed. Embryo vitrification was performed according to the Cryotop method (Kitazato, Japan) on day 5 or 6 of embryo development after a laser-induced collapse. Blastocysts were transferred from the culture medium (G2, Vitrolife) into an equilibration solution using 130 μm strippers. Blastomeres spontaneously began shrinking but gradually returned to their original size. This process took approximately 13 min. After that, blastocysts were transferred to a vitrification solution. Embryos were placed in VS1 for 30 s, VS2 for 30 s and then placed in the Cryotop sheet with minimal volume (<0.1 μL) of VS2. Excess volume was removed and the Cryotop was quickly plunged into fresh liquid nitrogen. Cryotops were stored in Espace 151 (Air Liquid, France) in the gas phase. Once a euploid embryo was available for transfer, endometrial preparation was induced with the sequential provision of oral estradiol valerate (Progyluton Bayer Hispania, Spain) in a fixed dose of 4 mg (per os) per day from cycle day 2 onward. After 8–10 days of estrogen therapy, a blood sample was collected, and a vaginal ultrasound was performed for measurements of progesterone levels (PP0) and endometrial features, respectively. Endometrial features included endometrial thickness (measured in the midsagittal plane of the uterus as the maximum distance between the two interfaces of endometrial–myometrial junction) and pattern ( Fig. 2 ). Treatment schedule. Pre-ET progesterone administration was started before the 14th day of the cycle and only if plasmatic progesterone determination (PP0) was lower than 1.0 ng/mL ( Coughlan et al. 2023 ). Based on the progesterone administration approach before embryo transfer (pre-ET progesterone), patients were divided into group A with natural micronized progesterone (Utrogestan, SEID, Spain) administered orally at a dose of 200 mg/8 h (600 mg/day) starting 6 days (in the morning) before scheduled embryo-transfer and group B with natural micronized progesterone vaginal administration (400 mg/12 h; 800 mg/day) starting (in the evening) 6 days before embryo transfer ( Fig. 2 ). Vitrified euploid blastocysts were warmed on the day of embryo transfer in those embryos vitrified on day 6 and the day before transfer in those vitrified on day 5, using thawing, equilibration and washing media and the protocol from the package insert (Irvine Scientific, USA). Single embryos were loaded into a transfer catheter (Sureview, Wallace, Smith’s Medical, UK) and transferred into the patient’s uterus. During the transfer procedure, the placement of the catheter within the uterus was confirmed by abdominal ultrasound. After embryo transfer, a vaginal protocol was administered (micronized progesterone, 800 mg/day, Utrogestan SEID, Spain). After an initial plasmatic progesterone determination to rule out patients before exogenous progesterone administration (PP0) (see above), two additional progesterone concentration determinations were included in the trial protocol. An analysis was performed on the day of embryo transfer (PP1). Finally, a new analysis was performed on day of B-HCG determination (10 days after ET) (PP2). Blood samples for progesterone concentration measurements were performed at 8:00 and 10:00 h. Levels of progesterone were analyzed by time-resolved fluoroimmunoassay (Roche® G3 Elecsys cobas e411, Switzerland), with a coefficient of variation of 0.10%. The detection limit was 0.05 ng/mL. If B-HCG determination was positive, hormonal treatment (including estrogen and progesterone) was maintained, and an ultrasound assessment was performed 2 weeks later. Once clinical pregnancy was diagnosed, patients continued with the prescribed hormonal supplementation until the 10th week of gestation. In the present study, the effects on birth outcomes also were investigated. All pregnant patients were followed until delivery. The primary outcome measure was to assess LBRs depending on the protocol of progesterone supplementation before embryo transfer (ET) and progesterone levels on the day of ET. LBR was defined as a viable intrauterine pregnancy resulting in the delivery of a viable infant after 24 weeks of gestation per cycle. The secondary outcomes included the clinical pregnancy, miscarriage rates and LBR depending on various factors, including the plasmatic progesterone levels on the day of B-HCG determination and endometrial thickness. Clinical pregnancy was defined as a visible gestational sac determined by vaginal ultrasonography performed 2 weeks after a positive pregnancy test (7 weeks of gestation). Miscarriage was defined as the loss of pregnancy within 14 weeks of confirmation of a clinical pregnancy.

Our findings reinforce the equivalence of oral and vaginal progesterone administration in FET cycles regarding clinical pregnancy and LBRs. While plasma progesterone levels on ET day were associated with pregnancy outcomes, other parameters, such as post-ET progesterone levels, showed no predictive value. These results emphasize the importance of personalized approaches to FET preparation and highlight areas for future investigation.

This randomized clinical trial comprehensively evaluates different progesterone administration schedules and plasma levels on the day of embryo transfer (ET) on full-term pregnancy rates in patients undergoing frozen-thawed single embryo transfer (FET) of euploid blastocysts. With the global increase in FET, unresolved debates persist. First, no consensus has been reached on the optimal regimen for endometrial preparation in these cycles despite advancements that leave significant uncertainty in strategies to enhance day-to-day clinical practice and success rates in FET ( Mackens et al. 2017 ). Moreover, focusing on (LBRs) and maternal, obstetrical and neonatal outcomes is crucial ( Hubayter & Muasher 2008 , Mumusoglu et al. 2021 ). Recent studies suggest that a well-conducted natural cycle is superior to an HRT cycle regarding obstetrical outcomes ( Melado Vidales et al. 2023 ). The optimal plasmatic progesterone levels around the day of ET for achieving satisfactory clinical outcomes after FET are debated, where the timing and concentration of progesterone exposure are crucial for establishing and maintaining a pregnancy ( Kofinas et al. 2015 , Devine et al. 2021). In addition, studies highlight conflicts over the best route for progesterone administration, reflecting differing opinions on oral versus vaginal routes ( Yarali et al. 2016 , Labarta et al. 2017 , Akaeda et al. 2019 ). The optimal route for progesterone replacement in FET remains unknown, with administration types including oral, vaginal, oil-based intramuscular (IM) and aqueous subcutaneous (SC) progesterone. Introducing non-intramuscular formulations provides physicians and patients with additional supplement options ( Aflatoonian & Mohammadi, 2021 ). Traditionally, vaginal and intramuscular routes have been the most effective and widely used ( Bulletti et al. 2022 , Zhu et al. 2023 ). Micronized progesterone shows enhanced absorption through both vaginal and intestinal mucosa; it is widely used for luteal phase supplementation in ART via the vaginal route, whereas oral progesterone is less commonly utilized ( Enatsu et al. 2018 , Melo et al. 2021 ). The vaginal route has become more popular due to reduced discomfort and fewer adverse effects, yielding high progesterone levels in the uterine endometrium, potentially favoring pregnancy outcomes ( Jiang et al. 2019 ). In our study, vaginal and oral progesterone administration before ET were evaluated, as these are commonly used alternatives to intramuscular and subcutaneous routes, assuming that oral administration post-ET is ineffective ( Saucedo et al. 2000 ). Our findings align with previous studies indicating no significant differences in live birth or clinical pregnancy rates between different protocols ( Ghobara & Vandekerckhove 2008 , Groenewoud et al. 2013 ). Despite lower progesterone concentrations with vaginal administration on the day of ET, clinical outcomes did not differ between treatment schedules. Our findings showed no difference in clinical outcomes, including live birth and miscarriage rates, between the two administration routes despite variations in plasma levels. Plasma progesterone levels are lower with the vaginal route compared to intramuscular ( Cicinelli et al. 2000 , Enatsu et al. 2018 ). However, all administration routes have been shown to achieve satisfactory mean serum progesterone levels by day 5 ( Labarta et al. 2021 , Melo et al. 2021 ). This trial identified a significant association between plasma progesterone levels on ET day and ongoing pregnancies. However, progesterone levels measured 10 days post-ET showed no correlation with pregnancy outcomes, suggesting that pre-implantation progesterone exposure plays a more critical role. The relationship between serum progesterone levels and pregnancy outcomes has been extensively studied, with mixed results ( Stavridis et al. 2023 ). Low serum progesterone levels have been linked to poor pregnancy outcomes in several studies ( Brady et al. 2014 , Yovich et al. 2015 , Labarta et al. 2017 , Cedrin-Durnerin et al. 2019 , Gaggiotti-Marre et al. 2019 , Alsbjerg et al. 2020 , Gonzalez-Foruria et al. 2020 , Labarta et al. 2021 , Shiba et al. 2021 ). However, the predictive value remains debated ( Melo et al. 2021 , Bulletti et al. 2022 , Lawrenz et al. 2023 , 2024 ). Although low progesterone levels on the ET day are associated with poor reproductive outcomes, the exact thresholds for optimal outcomes remain unclear ( Labarta et al. 2021 , González-Foruria et al. 2023 ). Previous studies assessing progesterone levels and pregnancy outcomes have methodological differences. Timing of serum progesterone measurement varied, with some measuring on ET day and others before or after ET. Some researchers propose additional progesterone supplementation for low serum levels to optimize luteal phase support ( Álvarez et al. 2021 , Labarta et al. 2022 ), but good-quality interventional evidence is lacking ( Alsbjerg et al. 2020 ). Although thin endometrium (<7 mm) is generally considered unfavorable, emerging evidence suggests that endometrial receptivity may depend more on functional characteristics than on thickness alone ( Liu et al. 2018 , Zhang et al. 2018 , Kai-Lun Hu et al. 2021 , Shakerian et al. 2021 , Ata et al. 2023 ). Our study showed that endometrial thickness was greater in patients with ongoing pregnancies than non-ongoing pregnancies. Paradoxically, endometrial thickness was also greater in non-pregnant patients than in those who experienced a miscarriage. This apparent contradiction highlights the need for a more nuanced understanding of endometrial receptivity markers. This trial exhibited a high cesarean section rate. HRT cycles are associated with increased risks of postpartum hemorrhage and cesarean section compared to natural cycles. FETs after hormonal replacement have a higher likelihood of hypertensive disorders and overall maternal complications ( Schattman 2023 ). This study contributes to the ongoing effort to optimize FET protocols by demonstrating that different progesterone administration regimens result in comparable clinical outcomes, despite differences in plasma levels. These findings support the use of patient-preferred routes of administration, particularly vaginal progesterone, which is associated with fewer side effects and greater convenience. Future research should focus on standardizing the timing and methods of progesterone measurement, as variations in study designs have led to inconsistent conclusions. In addition, well-designed interventional studies are needed to determine whether targeted progesterone supplementation for patients with low levels could improve outcomes.

The sample size calculation for this study was based on the rate of newborns as the primary outcome. The required sample size to detect a difference of 15% in the primary outcome measure (40%) ( Ferrando et al. 2020 ) with a two-tailed alpha of 0.05 and power (1-β) of 0.80 was estimated at 240 women in each arm. Data were expressed as the means and standard error of means (SEMs) for continuous variables or as frequency rates for categorical variables. The Student's t -test and one-way ANOVA for within-group variance analysis were used to compare continuous variables between the groups depending on the data distribution. Pearson’s chi-squared was used to compare categorical variables. Receiver operator characteristic (ROC) curves and the analysis of the area under the curve (AUC) were performed to evaluate the discriminatory ability of the assessed parameters to predict a successful clinical outcome. Serum progesterone levels on the day of embryo transfer were analyzed as both a continuous variable and a dichotomous variable (according to percentiles 10th, 25th and 50th). Statistical significance was set at a probability ( P ) value <0.05. Analyses were performed using the SPSS (IBM SPSS Statistics for Windows, version 29.0.2.0, released in 2023, IBM Corp., USA).

Successful embryo implantation is a process that requires synchronous development and relies upon crosstalk between the implanting euploid embryo at the blastocyst stage and a receptive endometrium ( Patel et al. 2019 , Casper 2020 ). Implantation failure of even euploid embryos is one of the most critical limiting factors in ART success and remains a ‘black hole’ in our knowledge. Endometrial receptivity can be compromised in fresh cycles by controlled ovarian hyperstimulation protocols and inadequate secretory endometrial transformation. However, endometrial development in frozen-thawed cycles (FET) mitigates the risk of endometrial asynchrony, allowing greater flexibility in the timing of embryo transfer ( Eftekhar et al. 2013 , Mackens et al. 2017 ). The utilization of FET treatment, particularly employing single-embryo transfer policies, significantly diminishes the likelihood of ovarian hyperstimulation syndrome and complications arising from multiple pregnancies linked with assisted reproductive procedures ( Wong et al. 2017 , Barrenetxea et al. 2020 , Melo et al. 2021 , Zhu et al. 2023 ). Since the first live birth by FET in 1983, these procedures have progressively advanced and become essential tools in the treatment of infertility ( Wei et al. 2019 ). Despite the worldwide increase in FET for various indications, the search for the best protocol to prepare the endometrium continues ( Mumusoglu et al. 2021 ), and there is still no consensus on which regimen is most optimal for endometrial preparation in FET cycles ( Ghobara & Vandekerckhove 2008 , Wangren et al. 2022 ). Hormonally substituted AC-FET requires less monitoring than other FET protocols, and thus offers flexibility in the timing of the thaw and transfer of the embryo ( Ortega & García Velasco 2015 ), although recent studies have shown that a correctly conducted natural cycle is superior to hormone replacement therapy (HRT) cycle regarding obstetrical outcome ( Melado Vidales et al. 2023 ). It has been reported that circulating progesterone during hormone replacement therapy in AC cycles may affect ongoing pregnancy in FET cycles ( Bulletti et al. 2022 ). Given the conflicting data on the optimal route of progesterone replacement for vitrified-warmed blastocyst transfer, highlighted by studies such as Akaeda et al. (2019) , Labarta et al. (2017) and Yarali et al. (2016) , which demonstrate existing conflicts on the best administration method, the increasing use of this treatment method and evidence of different preferences of patients, we conducted a trial to determine whether different progesterone replacement schedules before embryo transfer are better than others in terms of live birth rate (LBR) after transfer of vitrified euploid blastocysts. We also designed the trial to evaluate the potential influence of progesterone plasmatic levels around FET on a favorable clinical outcome.

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.

The data underlying this article will be shared by the corresponding author upon a reasonable request.

GB conceived the study idea. GB, OP, RC and JB participated in the design of the trial to recruit participants and assessed clinical outcomes. EM, MDLH, OG and OA were involved in the critical revision of the manuscript. GB and OP coordinated the data collection, performed the analysis and wrote the first draft of the manuscript. All authors approved the final manuscript.